August 19, 2012 at 10:03 pm
Austin FraktThe JAMA paper by Song, Cutler, and Chernew on the consequences of a Wyden-Ryan type premium support plan is getting a lot of attention. However, I’ve not seen anyone notice the following two points:
- Their 9% savings figure is an underestimate in the sense that it is measured relative to fee-for-service Medicare costs. Since Medicare Advantage plans are paid well above that level, the actual savings is more. I made this point in my post on the paper.
- Their 9% savings figure is an overestimate in the sense that it is a simulation of what would have happened had Wyden-Ryan been in effect in 2009. Current law includes the ACA, and relative to that, the savings would be less. Feldman, Coulam, and Dowd also examine premium support in 2009 and, using different methodology, come up with a 9.5% savings figure. However, when they factor in the effect of the ACA, their figure is cut to 5.6%.
It’s tempting to say 1 and 2 cancel each other out, and to some extent they do. It’d be nice to see someone work out exactly what the Song, Cutler, Chernew estimate would be taking into account the Medicare Advantage overpayments and the effect of the ACA. In other words, it’d be nice to see it relative to a current law baseline. As far as I know, nobody has made that estimate.
UPDATE: And then there’s the insufficient risk adjustment issue.
We find that differential payments actually rise after risk adjustment and estimate that they totaled $30 billion in 2006, or nearly eight percent of total Medicare spending.
That just about cancels any 9% saving.
by John Lynch on August 21st, 2012 at 13:42
The NBER study cites an risk-adjustement increase of 8% of total Medicare spending. Wouldn’t that EXCEED the 9% savings from the MA plans?
by Austin Frakt on August 21st, 2012 at 13:58
As my post, hopefully, makes clear, there’s a bit of apples-to-oranges going on and some omitted factors. I can’t conclude anything with precision from all this.